Changelog
What we shipped.
Releases, launches, and product news from the cognee team.
2026
Anthropic accidentally exposed 500,000 lines of internal source code. Buried inside: engineers studying Cognee's memory architecture and wondering if they should just adopt it.
Cognee raises $7.5M seed led by Pebblebed to build the open-source memory layer for AI agents—backed by founders of OpenAI and Facebook AI Research.
2025
Edge AI memory brings private, on-device AI memory to phones, wearables and IoT, for better latency, accuracy and privacy. Book a call to discuss your use case!
Temporal cognification brings time-awareness to AI memory, enabling LLMs to understand when events happened, how information evolved, and what context was relevant at any given moment.
Distributed cognee revolutionizes processing large datasets by enabling parallel execution on remote infrastructure, slashing processing times from 8+ hours to ~45 minutes while maintaining high-quality results.
Bridge cognee's AI memory and agent frameworks like LangGraph, OpenAI MCP, Anthropic MCP, and others with MCP (Model Context Protocol). Build smarter apps today!
Cognee September recap: AI memory upgrades, graph embeddings, new UI, and case studies. Improve context engineering and retrieval—read now and start building.
Enable role-based access control and dataset sharing in cognee to secure multi-tenant data across Kùzu and LanceDB via SDK or API. Manage AI memory permissions!
Build richer AI memory with weighted nodes and edges—enable memory layers and context engineering for recency and importance. Try Cognee for enhanced reasoning.
Cognee introduces graph-aware embeddings: graph signals boost semantic search for faster and more precise retrievals in paid plans. Learn more and book a call.
Simplify AI memory with cognee UI—notebooks, local + cloud sync, Graph Explorer, works with your databases. Deploy in seconds and start building smarter today!
AI memory gets smarter with real user feedback; auto-optimizing knowledge graphs boost search relevance and reduce manual tuning. See how it works and try it now!
Memify is a modular pipeline that keeps your knowledge graph evolving and speeds vector database updates for relevance and speed. Explore docs, join the beta now!
cognee August updates: Amazon Neptune integration, n8n for Cognee Cloud, community adapters, time-aware and self-improving graph, graph embeddings, and more!
cognee July updates: Launched Cognee Cloud, distributed processing, enhanced authentication, improved stability, case studies across construction, education, more!
AI security meets open source AI memory as cognee graduates GitHub Secure Open Source, proving enterprise grade practices for AI memory. Join us—try cognee today!
Explore cognee's June releases: Community vs Core split, Qdrant vector search integration, secure AI memory permissions, and SaaS beta waitlist. Join now!
Explore cognee's May 2025 updates with an agentic AI memory demo, academic paper, stability boosts, new UI, refreshed docs and more — try the features today!
Ontology, effortless relational database imports, hyperparameter tuning, major performance boosts... Enhance LLM grounding and AI graph capabilities with cognee!
2024
Discover how fundraising in 2024 in the AI infrastructure space looks like and how we went about securing 1.5m in funding. Join us in our journey growing cognee!
We're excited to announce that cognee has raised $1.5 million in funding! 🎉 Our mission is to make AI data management simpler, cost-effective, and scalable for you.
Earlier
Learn about differences between episodic and semantic memory systems, real-world use cases, challenges. In the AI Engineering Podcast, explore cognee's AI memory.